DAY 2

library(tidyverse)

Error in exists(cacheKey, where = .rs.WorkingDataEnv, inherits = FALSE) : 
  invalid first argument

library(janitor)

Error in exists(cacheKey, where = .rs.WorkingDataEnv, inherits = FALSE) : 
  invalid first argument

steam <- read_csv("raw_data/steam_checkpoint.csv")

Rows: 26564 Columns: 22── Column specification ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Delimiter: ","
chr  (8): name, developer, publisher, multiplayer, categories, genres, steamspy_tags, owners
dbl  (8): appid, required_age, achievements, positive_ratings, negative_ratings, average_playtime, median_playtime, price
lgl  (5): free_to_play, virtual_reality_support, windows_support, mac_support, linux_support
date (1): release_date
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.Error in exists(cacheKey, where = .rs.WorkingDataEnv, inherits = FALSE) : 
  invalid first argument

Where was i?

games_pre_cleaned <- read_csv("raw_data/Cleaned Data 2 GVGS&R.csv")

Rows: 6894 Columns: 15── Column specification ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Delimiter: ","
chr  (5): Name, Genre, Publisher, Developer, Rating
dbl (10): Year_of_Release, NA_Sales, EU_Sales, JP_Sales, Other_Sales, Global_Sales, Critic_Score, Critic_Count, User_Score, User_Count
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.Error in exists(cacheKey, where = .rs.WorkingDataEnv, inherits = FALSE) : 
  invalid first argument
Error in assign(cacheKey, frame, .rs.CachedDataEnv) : 
  attempt to use zero-length variable name

games_pre_cleaned

How dare you do the work for me

games_raw <- read_csv("raw_data/Raw Data GVGS&R.csv")

Rows: 16719 Columns: 16── Column specification ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Delimiter: ","
chr (8): Name, Platform, Year_of_Release, Genre, Publisher, User_Score, Developer, Rating
dbl (8): NA_Sales, EU_Sales, JP_Sales, Other_Sales, Global_Sales, Critic_Score, Critic_Count, User_Count
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

games_raw %>% 
  filter(Name == "Minecraft")

Why does everyone have the same dataset

back to steam i guess

steam

steam %>% 
  distinct(genres)

Already have F2P covered by another column, so we can drop that in genres

steam %>% 
  mutate(genres = str_remove_all(genres, "Free to Play")) %>% 
  filter(free_to_play == TRUE)

Is this worth doing? Duplicated info doesn’t hurt i suppose - leave it for now

single player?

steam <- steam %>% 
  mutate(singleplayer = case_when(
    str_detect(categories, "Single-player") ~ TRUE,
    str_detect(steamspy_tags, "Singleplayer") ~ TRUE,
    TRUE ~ FALSE
  ),.before = multiplayer)

steam %>% 
  filter(free_to_play == TRUE) 

backloggd <- read_csv("clean_data/backloggd_clean.csv")

Rows: 1115 Columns: 15── Column specification ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Delimiter: ","
chr  (9): title, team, genre_tag, genre_tag_2, genre_tag_3, genre_tag_4, genre_tag_5, genre_tag_6, genre_tag_7
dbl  (5): years_since_release, rating, number_of_reviews, wishlist, plays
date (1): release_date
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

backloggd

games_raw <- games_raw %>% 
  clean_names()

# Platform isn't a genre - correcting
games_raw <- games_raw %>% 
  mutate(genre = case_when(
    genre == "Platform" ~ "Platformer",
    TRUE ~ genre
  ))

nintendo_published <- games_raw %>% 
  filter(publisher == "Nintendo")

genre_publisher_sales <- games_raw %>%  
  group_by(publisher, genre) %>% 
  summarise(sum(global_sales)) %>% 
  arrange(desc(`sum(global_sales)`))

`summarise()` has grouped output by 'publisher'. You can override using the `.groups` argument.

write_csv(genre_publisher_sales, "clean_data/genre_by_publisher.csv")

# Adding PEGI equivalent 
games_raw <- games_raw %>% 
  mutate(pegi_equivalent = case_when(
    esrp == "EC" ~ "3",
    esrp == "E" ~ "7",
    esrp == "E10+" ~ "12",
    esrp == "T" ~ "16",
    esrp == "M" ~ "18",
    esrp == "AO" ~ "18",
    esrp == "K-A" ~ "7",
    esrp == "RP" ~ "No Rating",
    TRUE ~ "No Rating Found"
  )) 
  

how does the sales for the top 3 genres look in each region?

What about trends? How did each genres sales adjust every 5 or so years?

sales_2019 <- read_csv("raw_data/sales-2019.csv")

Rows: 55792 Columns: 23── Column specification ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Delimiter: ","
chr (10): Name, basename, Genre, ESRB_Rating, Platform, Publisher, Developer, Last_Update, url, img_url
dbl (12): Rank, Critic_Score, User_Score, Total_Shipped, Global_Sales, NA_Sales, PAL_Sales, JP_Sales, Other_Sales, Year, status, Vgchartzscore
lgl  (1): VGChartz_Score
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

Going off the fact that there is the majority of usable information i am able to get is from 2019 or before, we’re just going to pretend we’re living in an

alternative 2019

As such, i am dropping all data from 2020 in this dataset

Looking at backloggd again

backloggd %>% 
  mutate(rating_range = case_when(
    rating < 5 & rating >= 4 ~ "4+",
    rating < 4 & rating >= 3 ~ "3 to 4",
    rating < 3 & rating >= 2 ~ "2 to 3",
    rating < 2 & rating >= 1 ~ "1 to 2",
    rating < 1 ~ ">1",
    TRUE ~ "No rating"
  ))